|
|
| |
| # lines: |
312 | | # code: |
312 | | # comment: | 0 | |
# blank: | 0 |
| # Variables: | 44 |
| # Callers: | 1 |
| # Callings: | 3 |
| # Words: | 116 |
| # Keywords: | 68 |
|
|
|
|
|
..
.. Array Arguments ..
..
Purpose
=======
PDGELQF computes a LQ factorization of a real distributed M-by-N
matrix sub( A ) = A(IA:IA+M-1,JA:JA+N-1) = L * Q.
Notes
=====
Each global data object is described by an associated description
vector. This vector stores the information required to establish
the mapping between an object element and its corresponding process
and memory location.
Let A be a generic term for any 2D block cyclicly distributed array.
Such a global array has an associated description vector DESCA.
In the following comments, the character _ should be read as
"of the global array".
NOTATION STORED IN EXPLANATION
--------------- -------------- --------------------------------------
DTYPE_A(global) DESCA( DTYPE_ )The descriptor type. In this case,
DTYPE_A = 1.
CTXT_A (global) DESCA( CTXT_ ) The BLACS context handle, indicating
the BLACS process grid A is distribu-
ted over. The context itself is glo-
bal, but the handle (the integer
value) may vary.
M_A (global) DESCA( M_ ) The number of rows in the global
array A.
N_A (global) DESCA( N_ ) The number of columns in the global
array A.
MB_A (global) DESCA( MB_ ) The blocking factor used to distribute
the rows of the array.
NB_A (global) DESCA( NB_ ) The blocking factor used to distribute
the columns of the array.
RSRC_A (global) DESCA( RSRC_ ) The process row over which the first
row of the array A is distributed.
CSRC_A (global) DESCA( CSRC_ ) The process column over which the
first column of the array A is
distributed.
LLD_A (local) DESCA( LLD_ ) The leading dimension of the local
array. LLD_A >= MAX(1,LOCr(M_A)).
Let K be the number of rows or columns of a distributed matrix,
and assume that its process grid has dimension p x q.
LOCr( K ) denotes the number of elements of K that a process
would receive if K were distributed over the p processes of its
process column.
Similarly, LOCc( K ) denotes the number of elements of K that a
process would receive if K were distributed over the q processes of
its process row.
The values of LOCr() and LOCc() may be determined via a call to the
ScaLAPACK tool function, NUMROC:
LOCr( M ) = NUMROC( M, MB_A, MYROW, RSRC_A, NPROW ),
LOCc( N ) = NUMROC( N, NB_A, MYCOL, CSRC_A, NPCOL ).
An upper bound for these quantities may be computed by:
LOCr( M ) <= ceil( ceil(M/MB_A)/NPROW )*MB_A
LOCc( N ) <= ceil( ceil(N/NB_A)/NPCOL )*NB_A
Arguments
=========
M (global input) INTEGER
The number of rows to be operated on, i.e. the number of rows
of the distributed submatrix sub( A ). M >= 0.
N (global input) INTEGER
The number of columns to be operated on, i.e. the number of
columns of the distributed submatrix sub( A ). N >= 0.
A (local input/local output) DOUBLE PRECISION pointer into the
local memory to an array of dimension (LLD_A, LOCc(JA+N-1)).
On entry, the local pieces of the M-by-N distributed matrix
sub( A ) which is to be factored. On exit, the elements on
and below the diagonal of sub( A ) contain the M by min(M,N)
lower trapezoidal matrix L (L is lower triangular if M <= N);
the elements above the diagonal, with the array TAU, repre-
sent the orthogonal matrix Q as a product of elementary
reflectors (see Further Details).
IA (global input) INTEGER
The row index in the global array A indicating the first
row of sub( A ).
JA (global input) INTEGER
The column index in the global array A indicating the
first column of sub( A ).
DESCA (global and local input) INTEGER array of dimension DLEN_.
The array descriptor for the distributed matrix A.
TAU (local output) DOUBLE PRECISION array, dimension
LOCr(IA+MIN(M,N)-1). This array contains the scalar factors
of the elementary reflectors. TAU is tied to the distributed
matrix A.
WORK (local workspace/local output) DOUBLE PRECISION array,
dimension (LWORK)
On exit, WORK(1) returns the minimal and optimal LWORK.
LWORK (local or global input) INTEGER
The dimension of the array WORK.
LWORK is local input and must be at least
LWORK >= MB_A * ( Mp0 + Nq0 + MB_A ), where
IROFF = MOD( IA-1, MB_A ), ICOFF = MOD( JA-1, NB_A ),
IAROW = INDXG2P( IA, MB_A, MYROW, RSRC_A, NPROW ),
IACOL = INDXG2P( JA, NB_A, MYCOL, CSRC_A, NPCOL ),
Mp0 = NUMROC( M+IROFF, MB_A, MYROW, IAROW, NPROW ),
Nq0 = NUMROC( N+ICOFF, NB_A, MYCOL, IACOL, NPCOL ),
and NUMROC, INDXG2P are ScaLAPACK tool functions;
MYROW, MYCOL, NPROW and NPCOL can be determined by calling
the subroutine BLACS_GRIDINFO.
If LWORK = -1, then LWORK is global input and a workspace
query is assumed; the routine only calculates the minimum
and optimal size for all work arrays. Each of these
values is returned in the first entry of the corresponding
work array, and no error message is issued by PXERBLA.
INFO (global output) INTEGER
= 0: successful exit
< 0: If the i-th argument is an array and the j-entry had
an illegal value, then INFO = -(i*100+j), if the i-th
argument is a scalar and had an illegal value, then
INFO = -i.
Further Details
===============
The matrix Q is represented as a product of elementary reflectors
Q = H(ia+k-1) H(ia+k-2) . . . H(ia), where k = min(m,n).
Each H(i) has the form
H(i) = I - tau * v * v'
where tau is a real scalar, and v is a real vector with v(1:i-1)=0
and v(i) = 1; v(i+1:n) is stored on exit in A(ia+i-1,ja+i:ja+n-1),
and tau in TAU(ia+i-1).
=====================================================================
.. Parameters ..
|
|
|
|
001 SUBROUTINE PDGELQF( M , N , A , IA , JA , DESCA , TAU , WORK , LWORK ,
002 $INFO )
003
004 * -- ScaLAPACK routine(version 1.7) --
005 * University of Tennessee , Knoxville , Oak Ridge National Laboratory ,
006 * and University of California , Berkeley.
007 * May 25 , 2001
008
009 * .. Scalar Arguments ..
010 INTEGER IA , INFO , JA , LWORK , M , N
011 INTEGER BLOCK_CYCLIC_2D , CSRC_ , CTXT_ , DLEN_ , DTYPE_ ,
012 $LLD_ , MB_ , M_ , NB_ , N_ , RSRC_
013 PARAMETER( BLOCK_CYCLIC_2D = 1 , DLEN_ = 9 , DTYPE_ = 1 ,
014 $CTXT_ = 2 , M_ = 3 , N_ = 4 , MB_ = 5 , NB_ = 6 ,
015 $RSRC_ = 7 , CSRC_ = 8 , LLD_ = 9 )
016 * ..
017 * .. Local Scalars ..
018 LOGICAL LQUERY
019 CHARACTER COLBTOP , ROWBTOP
020 INTEGER I , IACOL , IAROW , IB , ICTXT , IINFO , IN , IPW ,
021 $IROFF , J , K , LWMIN , MP0 , MYCOL , MYROW , NPCOL ,
022 $NPROW , NQ0
023 * ..
024 * .. Local Arrays ..
025 INTEGER IDUM1( 1 ) , IDUM2( 1 )
026 * ..
027 * .. External Subroutines ..
028 EXTERNAL BLACS_GRIDINFO , CHK1MAT , PCHK1MAT , PDGELQ2 ,
029 $PDLARFB , PDLARFT , PB_TOPGET , PB_TOPSET , PXERBLA
030 * ..
031 * .. External Functions ..
032 INTEGER ICEIL , INDXG2P , NUMROC
033 EXTERNAL ICEIL , INDXG2P , NUMROC
034 * ..
035 * .. Intrinsic Functions ..
036 INTRINSIC DBLE , MIN , MOD
037 * ..
038 * .. Executable Statements ..
039
040 * Get grid parameters
041
042 ICTXT = DESCA( CTXT_ )
043 CALL BLACS_GRIDINFO( ICTXT , NPROW , NPCOL , MYROW , MYCOL )
044
045 * Test the input parameters
046
047 INFO = 0
048 IF( NPROW.EQ. - 1 ) THEN
048
049 INFO = - (600 + CTXT_)
050 ELSE
050
051 CALL CHK1MAT( M , 1 , N , 2 , IA , JA , DESCA , 6 , INFO )
052 IF( INFO.EQ.0 ) THEN
052
053 IROFF = MOD( IA - 1 , DESCA( MB_ ) )
054 IAROW = INDXG2P( IA , DESCA( MB_ ) , MYROW , DESCA( RSRC_ ) ,
055 $ NPROW )
056 IACOL = INDXG2P( JA , DESCA( NB_ ) , MYCOL , DESCA( CSRC_ ) ,
057 $ NPCOL )
058 MP0 = NUMROC( M + IROFF , DESCA( MB_ ) , MYROW , IAROW , NPROW )
059 NQ0 = NUMROC( N + MOD( JA - 1 , DESCA( NB_ ) ) , DESCA( NB_ ) ,
060 $ MYCOL , IACOL , NPCOL )
061 LWMIN = DESCA( MB_ ) * ( MP0 + NQ0 + DESCA( MB_ ) )
062
063 WORK( 1 ) = DBLE( LWMIN )
064 LQUERY =( LWORK.EQ. - 1 )
065 IF( LWORK.LT.LWMIN .AND. .NOT.LQUERY )
065
066 $ INFO = - 9
067 END IF
068 IF( LWORK.EQ. - 1 ) THEN
068
069 IDUM1( 1 ) = - 1
070 ELSE
070
071 IDUM1( 1 ) = 1
072 END IF
073 IDUM2( 1 ) = 9
074 CALL PCHK1MAT( M , 1 , N , 2 , IA , JA , DESCA , 6 , 1 , IDUM1 , IDUM2 ,
075 $ INFO )
076 END IF
077
078 IF( INFO.NE.0 ) THEN
078
079 CALL PXERBLA( ICTXT , 'PDGELQF' , - INFO )
080 RETURN
081 ELSE IF( LQUERY ) THEN
081
082 RETURN
083 END IF
084
085 * Quick return if possible
086
087 IF( M.EQ.0 .OR. N.EQ.0 )
087
088 $ RETURN
089
090 K = MIN( M , N )
091 IPW = DESCA( MB_ ) * DESCA( MB_ ) + 1
092 CALL PB_TOPGET( ICTXT , 'Broadcast' , 'Rowwise' , ROWBTOP )
093 CALL PB_TOPGET( ICTXT , 'Broadcast' , 'Columnwise' , COLBTOP )
094 CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Rowwise' , ' ' )
095 CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Columnwise' , 'I - ring' )
096
097 * Handle the first block of rows separately
098
099 IN = MIN( ICEIL( IA , DESCA( MB_ ) ) * DESCA( MB_ ) , IA + K - 1 )
100 IB = IN - IA + 1
101
102 * Compute the LQ factorization of the first block A(ia : in : ja : ja + n - 1)
103
104 CALL PDGELQ2 ( IB , N , A , IA , JA , DESCA , TAU , WORK , LWORK , IINFO )
105
106 IF( IA + IB.LE.IA + M - 1 ) THEN
107
108 * Form the triangular factor of the block reflector
109 * H = H(ia) H(ia + 1) . . . H(in)
110
110
111 CALL PDLARFT ( 'Forward' , 'Rowwise' , N , IB , A , IA , JA , DESCA ,
112 $ TAU , WORK , WORK( IPW ) )
113
114 * Apply H to A(ia + ib : ia + m - 1 , ja : ja + n - 1) from the right
115
116 CALL PDLARFB ( 'Right' , 'No transpose' , 'Forward' , 'Rowwise' ,
117 $ M - IB , N , IB , A , IA , JA , DESCA , WORK , A , IA + IB ,
118 $ JA , DESCA , WORK( IPW ) )
119 END IF
120
121 * Loop over the remaining blocks of rows
122
123 DO 10 I = IN + 1 , IA + K - 1 , DESCA( MB_ )
123
124 IB = MIN( K - I + IA , DESCA( MB_ ) )
125 J = JA + I - IA
126
127 * Compute the LQ factorization of the current block
128 * A(i : i + ib - 1 : j : ja + n - 1)
129
130 CALL PDGELQ2 ( IB , N - I + IA , A , I , J , DESCA , TAU , WORK , LWORK ,
131 $ IINFO )
132
133 IF( I + IB.LE.IA + M - 1 ) THEN
134
135 * Form the triangular factor of the block reflector
136 * H = H(i) H(i + 1) . . . H(i + ib - 1)
137
137
138 CALL PDLARFT ( 'Forward' , 'Rowwise' , N - I + IA , IB , A , I , J ,
139 $ DESCA , TAU , WORK , WORK( IPW ) )
140
141 * Apply H to A(i + ib : ia + m - 1 , j : ja + n - 1) from the right
142
143 CALL PDLARFB ( 'Right' , 'No transpose' , 'Forward' , 'Rowwise' ,
144 $ M - I - IB + IA , N - J + JA , IB , A , I , J , DESCA , WORK ,
145 $ A , I + IB , J , DESCA , WORK( IPW ) )
146 END IF
147
148 10 CONTINUE
149
149
150 CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Rowwise' , ROWBTOP )
151 CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Columnwise' , COLBTOP )
152
153 WORK( 1 ) = DBLE( LWMIN )
154
155 RETURN
156
157 * End of PDGELQF
158
159 END35
13
|
|
Variables in Routine PDGELQF()
| Summary Report |
| Data Type | Quantity | Size(byte) |
| CHARACTER | 2 | 2 |
| INTEGER | 40 | 164 |
| LOGICAL | 1 | 1 |
| REAL | 1 | 4 |
| TOTAL | 44 | 171 |
List of Variables
CHARACTER
INTEGER
| BLOCK_CYCLIC_2D | CSRC_ | CTXT_ | DLEN_ | DTYPE_ |
| I | IA | IACOL | IAROW | IB |
| ICEIL | ICTXT | IDUM1( 1 ) | IDUM2( 1 ) | IINFO |
| IN | INDXG2P | INFO | IPW | IROFF |
| J | JA | K | LLD_ | LWMIN |
| LWORK | M | M_ | MB_ | MP0 |
| MYCOL | MYROW | N | N_ | NB_ |
| NPCOL | NPROW | NQ0 | NUMROC | RSRC_ |
LOGICAL
REAL
Variables Dependence Graph Put the mouse over a right hand side variable to display the corresponding line of the dependence | | - | | - | - | | I | <--- | INDO 10 I = IN+1, IA+K-1, DESCA( MB_ ), KDO 10 I = IN+1, IA+K-1, DESCA( MB_ ), MB_DO 10 I = IN+1, IA+K-1, DESCA( MB_ ), IADO 10 I = IN+1, IA+K-1, DESCA( MB_ ) |
| IACOL | <--- | INDXG2PIACOL = INDXG2P( JA, DESCA( NB_ ), MYCOL, DESCA( CSRC_ ),, JAIACOL = INDXG2P( JA, DESCA( NB_ ), MYCOL, DESCA( CSRC_ ),, CSRC_IACOL = INDXG2P( JA, DESCA( NB_ ), MYCOL, DESCA( CSRC_ ),, MYCOLIACOL = INDXG2P( JA, DESCA( NB_ ), MYCOL, DESCA( CSRC_ ),, NB_IACOL = INDXG2P( JA, DESCA( NB_ ), MYCOL, DESCA( CSRC_ ),, NPCOLIACOL = INDXG2P( JA, DESCA( NB_ ), MYCOL, DESCA( CSRC_ ), |
| IAROW | <--- | INDXG2PIAROW = INDXG2P( IA, DESCA( MB_ ), MYROW, DESCA( RSRC_ ),, MB_IAROW = INDXG2P( IA, DESCA( MB_ ), MYROW, DESCA( RSRC_ ),, MYROWIAROW = INDXG2P( IA, DESCA( MB_ ), MYROW, DESCA( RSRC_ ),, NPROWIAROW = INDXG2P( IA, DESCA( MB_ ), MYROW, DESCA( RSRC_ ),, RSRC_IAROW = INDXG2P( IA, DESCA( MB_ ), MYROW, DESCA( RSRC_ ),, IAIAROW = INDXG2P( IA, DESCA( MB_ ), MYROW, DESCA( RSRC_ ), |
| IB | <--- | INIB = IN - IA + 1, KIB = MIN( K-I+IA, DESCA( MB_ ) ), MB_IB = MIN( K-I+IA, DESCA( MB_ ) ), IIB = MIN( K-I+IA, DESCA( MB_ ) ), IAIB = IN - IA + 1{2IB = MIN( K-I+IA, DESCA( MB_ ) )} |
| ICTXT | <--- | CTXT_ICTXT = DESCA( CTXT_ ) |
| IN | <--- | ICEILIN = MIN( ICEIL( IA, DESCA( MB_ ) ) * DESCA( MB_ ), IA+K-1 ), KIN = MIN( ICEIL( IA, DESCA( MB_ ) ) * DESCA( MB_ ), IA+K-1 ), MB_IN = MIN( ICEIL( IA, DESCA( MB_ ) ) * DESCA( MB_ ), IA+K-1 ), IAIN = MIN( ICEIL( IA, DESCA( MB_ ) ) * DESCA( MB_ ), IA+K-1 ) |
| INFO | <--- | CTXT_INFO = -(600+CTXT_) |
| IPW | <--- | MB_IPW = DESCA( MB_ ) * DESCA( MB_ ) + 1 |
| IROFF | <--- | MB_IROFF = MOD( IA-1, DESCA( MB_ ) ), IAIROFF = MOD( IA-1, DESCA( MB_ ) ) |
| J | <--- | JAJ = JA + I - IA, IJ = JA + I - IA, IAJ = JA + I - IA |
| K | <--- | MK = MIN( M, N ), NK = MIN( M, N ) |
| LWMIN | <--- | MB_LWMIN = DESCA( MB_ ) * ( MP0 + NQ0 + DESCA( MB_ ) ), MP0LWMIN = DESCA( MB_ ) * ( MP0 + NQ0 + DESCA( MB_ ) ), NQ0LWMIN = DESCA( MB_ ) * ( MP0 + NQ0 + DESCA( MB_ ) ) |
| MP0 | <--- | IAROWMP0 = NUMROC( M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW ), IROFFMP0 = NUMROC( M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW ), MMP0 = NUMROC( M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW ), MB_MP0 = NUMROC( M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW ), MYROWMP0 = NUMROC( M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW ), NPROWMP0 = NUMROC( M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW ), NUMROCMP0 = NUMROC( M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW ) |
| NQ0 | <--- | JANQ0 = NUMROC( N+MOD( JA-1, DESCA( NB_ ) ), DESCA( NB_ ),, MYCOLNQ0 = NUMROC( N+MOD( JA-1, DESCA( NB_ ) ), DESCA( NB_ ),, NNQ0 = NUMROC( N+MOD( JA-1, DESCA( NB_ ) ), DESCA( NB_ ),, NB_NQ0 = NUMROC( N+MOD( JA-1, DESCA( NB_ ) ), DESCA( NB_ ),, NPCOLNQ0 = NUMROC( N+MOD( JA-1, DESCA( NB_ ) ), DESCA( NB_ ),, NUMROCNQ0 = NUMROC( N+MOD( JA-1, DESCA( NB_ ) ), DESCA( NB_ ),, IACOLNQ0 = NUMROC( N+MOD( JA-1, DESCA( NB_ ) ), DESCA( NB_ ), |
| WORK | <--- | LWMINWORK( 1 ) = DBLE( LWMIN ){2WORK( 1 ) = DBLE( LWMIN )} |
|
|
Analysis elements of the routine PDGELQF() Put the mouse over each element to display detailed matching information
Assigned variables |
| | | BLOCK_CYCLIC_2D , CSRC_ , CTXT_ , DLEN_ , DTYPE_ , I , IACOL , IAROW , IB , ICTXT , IDUM1 , IDUM2 , IN , INFO , IPW , IROFF , J , K , LLD_ , LQUERY , LWMIN , M_ , MB_ , MP0 , N_ , NB_ , NQ0 , RSRC_ , WORK |
|
Active variables |
| | | A , BLOCK_CYCLIC_2D , COLBTOP , CSRC_ , CTXT_ , DESCA , DLEN_ , DTYPE_ , I , IA , IACOL , IAROW , IB , ICEIL , ICTXT , IDUM1 , IDUM2 , IINFO , IN , INDXG2P , INFO , IPW , IROFF , J , JA , K , LLD_ , LQUERY , LWMIN , LWORK , M , M_ , MB_ , MP0 , MYCOL , MYROW , N , N_ , NB_ , NPCOL , NPROW , NQ0 , NUMROC , ROWBTOP , RSRC_ , TAU , WORK |
|
Accessed arrays [ array name : associated index ] |
| | A | : i:i+ib-1:j:ja+n-1 , i+ib:ia+m-1,j:ja+n-1 , ia:in:ja:ja+n-1 , ia+ib:ia+m-1,ja:ja+n-1 |
| | DESCA | : CSRC_ , CTXT_ , MB_ , MB_ , MB_ , MB_ , MB_ , MB_ , MB_ , MB_ , NB_ , NB_ , RSRC_ |
| | ICEIL | : IA, DESCA( MB_ ) |
| | IDUM1 | : 1 , 1 , 1 |
| | IDUM2 | : 1 , 1 |
| | NUMROC | : M+IROFF, DESCA( MB_ ), MYROW, IAROW, NPROW |
| | WORK | : 1 , 1 , IPW , IPW , IPW , IPW |
|
Conditional statements [ statement : associated predicate ] |
| | do | : ( 10 I = IN + 1 , IA + K - 1 , DESCA( MB_ ) ) |
| | if | : ( NPROW.EQ. - 1 ) , ( INFO.EQ.0 ) , ( LWORK.LT.LWMIN .AND. .NOT.LQUERY ) , ( LWORK.EQ. - 1 ) , ( INFO.NE.0 ) , ( LQUERY ) , ( possible ) , ( M.EQ.0 .OR. N.EQ.0 ) , ( IA+IB.LE.IA + M - 1 ) , ( I+IB.LE.IA + M - 1 ) |
|
| List of variables | BLOCK_CYCLIC_2D COLBTOP CSRC_ CTXT_ DLEN_ DTYPE_ I
| IA IACOL IAROW IB ICEIL ICTXT IDUM1( 1 ) IDUM2( 1 )
| IINFO IN INDXG2P INFO IPW IROFF J JA
| K LLD_ LQUERY LWMIN LWORK M M_ MB_
| MP0 MYCOL MYROW N N_ NB_ NPCOL NPROW
| NQ0 NUMROC ROWBTOP RSRC_ WORK | | close
| |
BLOCK_CYCLIC_2D
COLBTOP
CSRC_
CTXT_
DLEN_
DTYPE_
I
IA
IACOL
IAROW
IB
ICEIL
ICTXT
IDUM1( 1 )
IDUM2( 1 )
IINFO
IN
INDXG2P
INFO
IPW
IROFF
J
JA
K
LLD_
LQUERY
LWMIN
LWORK
M
M_
MB_
MP0
MYCOL
MYROW
N
N_
NB_
NPCOL
NPROW
NQ0
NUMROC
ROWBTOP
RSRC_
WORK
182#235#233
| |